After declining from the early 1980's until 1992, illicit drug use among adolescents has begun to increase. It is argued that adolescent substance use is almost a rite of passage and that adolescents will outgrow it. While some adolescents do stop on their own, most that start using marijuana regularly at an early age have been found to continue or increase their use. This trend is likely to worsen because the age of onset has been decreasing over the past 30 years. Adolescent admissions into treatment programs grew 53 percent from the years 1992 to 1998. The primary substance for the admission shifted from alcohol to marijuana during this time. Most of these treatment programs were using adult treatment models with only minimum modifications for adolescents. Few formal adolescent treatment models exist that have demonstrated effectiveness and affordability in community-based programs. The Adolescent Treatment Model (ATM) program was established to identify current exemplary models of adolescent treatment; formalize models into manuals to be replicated; determine the amount of services the adolescents received; and evaluate the effectiveness and cost of each model. The treatment models include outpatient treatment programs, outpatient family therapy, residential programs, therapeutic community treatment programs, and step-down treatment programs. Participants in all programs are predominately marijuana and alcohol using males with criminal justice system involvement. The Treatment Transition Log (TTL) is designed to track adolescents as they move through the ATM program. The Follow-Up Log (FUL) tracks all people recruited into the study, whether they agreed to participate, and, if so what happened to them. The Global Appraisal of Individual Needs (GAIN) is a series of screeners, intake, and follow-up participant interviews. The cost study is being conducted using an accounting methodology based on generally accepted accounting principles and cost accounting standards. 3 tables, 105 references